Popular repositories Loading

1. see-but-cant-say see-but-cant-say Public

   A Claude skill for when you can see the destination but can't describe the path

   1

2. fresh-eyes-review fresh-eyes-review Public

   A Claude Code skill that simulates real human readers (QA / dev / tech lead) reading your handover & onboarding docs — to catch what makes the team bounce it back, before you send it.

   1

3. handover-docs handover-docs Public

   Claude Code skill that writes reader-first handover / onboarding / handbook docs a real person can actually use. Companion to fresh-eyes-review.

4. poke-holes poke-holes Public

   Adversarial multi-persona review skill for Claude Code - finds the holes in a PoC / design / decision (auto council or due-diligence mode).

5. differing-view differing-view Public

   Independent epistemic critique skill for Claude Code - reviews knowledge artifacts from another AI agent (provenance, evidence independence, attribution drift) and writes a ranked differing-view as…

6. skill-shape-assessment skill-shape-assessment Public

   Evidence-based skill shape assessment skill for Claude Code - scores a person's skills (pure / +AI / false depth / upside, meta-soft-hard) from real work records, with bootstrap and blind re-assess…